S100 proteins are a family of Ca2+ binding EF-hand proteins. S100A8 is a cytosolic protein expressed in myeloid cells and epithelia where it forms a stable heterodimer with another S100 protein family member, S100A9. The S100A9 null mouse is viable and has no gross defect whereas the S100A8 null mouse is embryonic lethal. It was originally proposed that the S100A8 null mouse is lethal at E 9.0 in development due to lack of expression at E 6.5 in ectoplacental cone cells. This thesis shows that the S100A8 null phenotype is more complex than originally thought. S100A8 has a role in preimplantation development, which is previously unstudied. A small number of S100A8 null embryos survive to blastocyst but none survive implantation showing fatal compromise of S100A8 null embryos early in development. This thesis presents evidence that this lethality presents between fertilisation and E 2.5 of development. S100A8 also has a role in the murine decidua after implantation possibly key to normal murine development. S100A8 mRNA is highly expressed in maternal decidua yet S100A8 protein is not highly expressed. Foetal yolk sac cells do not express S100A8 mRNA yet they do stain positively for S100A8 protein. This thesis proposes that S100A8 protein is generated in the murine decidua and exported to the foetus where haematopoietic cells present the protein. The S100A8 protein has been shown to be expressed and stable independently of its myeloid partner, S100A9. These observations explain the discrepancy between the two SI00 null mouse phenotypes and add new insight to the S100A8 null phenotype.

The neural crest (NC) is an embryonic population of cells, which delaminate from the neural tube epithelium to become vigorous migratory cells that colonise the entire embryo and give rise to many different derivatives. Neural crest migration requires activation of the non-canonical Wnt/planar cell polarity (PCP) signalling pathway, but it is not known exactly how this pathway controls cell migration. Here I show that the PCP ligand, Wnt11R, and the downsteam PCP element, Dishevelled, are essential for neural crest migration in Xenopus laevis embryos. Additionally, the proteoglycan, Syndecan-4, interacts with Dishevelled to control NC migration. A detailed examination of neural crest cell behaviour in Xenopus and zebrafish embryos shows that, in the absence of Dishevelled or Syndecan-4, cells are motile but lack the persistent migration that allows them to reach their target tissue. Furthermore, Dishevelled and Syndecan-4 control directional migration by regulating the polarised formation of cell protrusions. They also regulate the formation of paxillin-containing focal contacts in vitro and in vivo. Rho GTPase activity was measured using FRET analysis in neural crest cells migrating in vitro and in vivo after interfering with Syndecan-4/PCP signalling. I demonstrate that Syndecan-4 acts as a potent inhibitor of Rac, while Dishevelled activates RhoA. In addition, I show that RhoA inhibits Rac in neural crest cells. So, modulation of Rac by Syndecan-4 and PCP signalling allows the polarised formation of cell protrusions required for persistent NC migration. Finally I show that cell-cell contact inhibition of locomotion, dependent on PCP signalling, contributes to the initial polarity of the cell by inhibiting cell protrusions. Thus I present a model whereby neural crest cells are able to establish and maintain a directed migration by the integration of signals from cell-cell interactions mediated by PCP signalling and from the extracellular matrix via Syndecan-4.

Bone Morphogenetic Proteins (BMPs) are members of the Transforming Growth Factor-beta (TGF-beta) signalling protein superfamily. BMPs play important and diverse roles in cell-cell signalling, including establishing cell fate during the development of vertebrate embryos. Their activity is antagonised in vivo by a number of proteins such as noggin, which sequester BMP ligands, preventing them from binding to BMP receptors. This thesis describes studies to establish a binary genetic approach combined with a ligand trap system to manipulate BMP signalling in the frog embryo. This system has been used to investigate the roles of BMP signalling in dorso-ventral patterning of the forebrain Xenopus tropicalis. The binary system described utilises a variety of tissue- or region-specific gene promoters to drive expression of the GAL4 transcriptional activator. Such transgenic "driver" lines can be crossed with a "responder" line in which expression of a membrane-tethered fusion protein comprising human Noggin fused to GFP is regulated by a synthetic promoter responsive to GAL4 (UAS-flognog). Transient expression assays confirmed the effectiveness of the "responder" line, GAL4 transactivation of UAS-flognog resulted in the expression of Flognog and an expansion of neural progenitor tissue, indicated by the X-Sox3 marker. In a binary cross with the Otx2-gal4 driver line, targeted GAL4 transactivation lead to a decrease in phospho-Smad-1 staining in the anterior CNS and eye in a proportion of cross embryos. Such a cross resulted in embryos showing an open neural tube and alterations in both Pax6 (dorsal) and X-<U13 (ventral) forebrain markers, further indicating the efficacy of the binary, ligand-trap strategy. In order to achieve temporal control on the activity of the UAS-flognog responder line in the telencephalon, an inducible driver line comprising the Pax6 promoter driving hormone-inducible GalPR (an inducible chimeric GAL4) was created. In binary crosses with a UAS-gfp reporter line, GFP expression was detected in the forebrain, hindbrain and spinal cord only in the presence of the steroid hormone, RU486. Similarly, a second driver line, N-tubulin-GalPR yielded inducible GFP expression in the developing brain, spinal cord and lens tissue in the presence of RU486. In conclusion, these findings are evidence that the binary ligand trap approach is functional and can cause targeted knockdown of BMP signalling, resulting in alterations in neural development and patterning. Furthermore, using an inducible version of this approach, Flognog (or any other target gene) can be expressed in the telencephalon in a RU486-inducible manner.

The epicardium is the outermost layer of the mature vertebrate heart. In birds and mice the epicardium originates from an extracardiac primordium - the proepicardium - and envelops the developing heart of the embryo. The epicardium of birds and mice has two main functions in normal heart development: 1) a subset of epicardial cells undergo an epithelial to mesenchymal transition, generating the pluripotent progenitors of the coronary vascular system 2) the epicardium secretes signals required for the proliferation and survival of cardiac myocytes. Whilst the first role has been the subject of considerable study, much less is understood about the role of the epicardium in maintenance and development of the heart myocardium. The heart of the amphibian, Xenopus laevis, differs from that of mammals in a number of ways. The frog myocardium has a spongy, less compact structure and completely lacks coronary vasculature. As a result, studying the epicardium in Xenopus could provide a useful model for investigating the role of this tissue in the structure and development of heart muscle. The structure of the epicardium in adult Xenopus laevis was analysed using histological methods and is shown to comprise a single cell-layered simple squamous epithelium. The embryonic origin and developmental progression of the epicardium was analysed histologically, by scanning electron microscopy, and by episcopic fluorescence image capturing. The epicardium is shown to be derived from the proepicardium, a clump of cells approximately 161 urn wide by 133pm high, located caudally to the heart on the septum transversum. The proepicardium is shown to be located asymmetrically to the right of the sagittal plane. Cell lineage analyses confirm the unilateral, right-sided origin of the epicardium. At around stage 42 in development the proepicardium can be seen to be in direct contact with the ventricular myocardium, and proepicardial cells can be observed migrating over this 'proepicardial bridge.' Once proepicardial cells contact the ventricular myocardium, they proceed to envelop the heart beginning with the atrioventricular and conoventricular canals. Epicardial cells in Xenopus laevis were identified using the expression patterns of three genes - Tbxl8, epicardin and Wilms tumour 1 - that have been shown to demarcate the avian or mammalian epicardium. Morpholino knock downs of these candidate genes were performed in order to investigate the roles of these genes in epicardium development, and therefore the wider roles of the epicardium in heart development. However, epicardium formation was found to be occurring too late in development to be accessible to morpholino knockdown analysis.

No description available.

No description available.

No description available.

No description available.

No description available.

No description available.